Matches in SemOpenAlex for { <https://semopenalex.org/work/W4365131790> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4365131790 endingPage "1" @default.
- W4365131790 startingPage "1" @default.
- W4365131790 abstract "Soft errors in large VLSI circuits have a significant impact on computing-and memory-intensive neural network (NN) processing. Understanding the influence of soft errors on NNs is critical to protect against soft errors for reliable NN processing. Prior work mainly rely on fault simulation to analyze the influence of soft errors on NN processing. They are accurate but usually specific to limited configurations of errors and NN models due to the prohibitively slow simulation speed especially for large NN models and datasets. With the observation that the influence of soft errors propagates across a large number of neurons and accumulates as well, we propose to characterize the soft error induced data disturbance on each neuron with normal distribution model using the central limit theorem and develop a series of statistical models to analyze the behavior of NN models under soft errors in general. The statistical models reveal not only the correlation between soft errors and the accuracy of NN models, but also how NN parameters such as quantization and architecture affect the reliability of NNs. The proposed models are compared with fault simulations and verified comprehensively. In addition, we observe that the statistical models that characterize the soft error influence can also be utilized to predict fault simulation results in many cases and we explore the use of the proposed statistical models to accelerate fault simulations of NNs. Our experiments show that the proposed accelerated fault simulation provides almost two orders of magnitude speedup with negligible loss of simulation accuracy compared to the baseline fault simulations." @default.
- W4365131790 created "2023-04-13" @default.
- W4365131790 creator A5002433945 @default.
- W4365131790 creator A5017317271 @default.
- W4365131790 creator A5021300680 @default.
- W4365131790 creator A5022364292 @default.
- W4365131790 creator A5023380073 @default.
- W4365131790 creator A5027061752 @default.
- W4365131790 creator A5050174081 @default.
- W4365131790 creator A5073321754 @default.
- W4365131790 date "2023-01-01" @default.
- W4365131790 modified "2023-09-24" @default.
- W4365131790 title "Statistical Modeling of Soft Error Influence on Neural Networks" @default.
- W4365131790 doi "https://doi.org/10.1109/tcad.2023.3266405" @default.
- W4365131790 hasPublicationYear "2023" @default.
- W4365131790 type Work @default.
- W4365131790 citedByCount "0" @default.
- W4365131790 crossrefType "journal-article" @default.
- W4365131790 hasAuthorship W4365131790A5002433945 @default.
- W4365131790 hasAuthorship W4365131790A5017317271 @default.
- W4365131790 hasAuthorship W4365131790A5021300680 @default.
- W4365131790 hasAuthorship W4365131790A5022364292 @default.
- W4365131790 hasAuthorship W4365131790A5023380073 @default.
- W4365131790 hasAuthorship W4365131790A5027061752 @default.
- W4365131790 hasAuthorship W4365131790A5050174081 @default.
- W4365131790 hasAuthorship W4365131790A5073321754 @default.
- W4365131790 hasConcept C11413529 @default.
- W4365131790 hasConcept C114289077 @default.
- W4365131790 hasConcept C121332964 @default.
- W4365131790 hasConcept C127313418 @default.
- W4365131790 hasConcept C127413603 @default.
- W4365131790 hasConcept C140073362 @default.
- W4365131790 hasConcept C154474529 @default.
- W4365131790 hasConcept C154945302 @default.
- W4365131790 hasConcept C163258240 @default.
- W4365131790 hasConcept C165205528 @default.
- W4365131790 hasConcept C173608175 @default.
- W4365131790 hasConcept C175551986 @default.
- W4365131790 hasConcept C24326235 @default.
- W4365131790 hasConcept C28855332 @default.
- W4365131790 hasConcept C41008148 @default.
- W4365131790 hasConcept C43214815 @default.
- W4365131790 hasConcept C50644808 @default.
- W4365131790 hasConcept C62520636 @default.
- W4365131790 hasConcept C68339613 @default.
- W4365131790 hasConceptScore W4365131790C11413529 @default.
- W4365131790 hasConceptScore W4365131790C114289077 @default.
- W4365131790 hasConceptScore W4365131790C121332964 @default.
- W4365131790 hasConceptScore W4365131790C127313418 @default.
- W4365131790 hasConceptScore W4365131790C127413603 @default.
- W4365131790 hasConceptScore W4365131790C140073362 @default.
- W4365131790 hasConceptScore W4365131790C154474529 @default.
- W4365131790 hasConceptScore W4365131790C154945302 @default.
- W4365131790 hasConceptScore W4365131790C163258240 @default.
- W4365131790 hasConceptScore W4365131790C165205528 @default.
- W4365131790 hasConceptScore W4365131790C173608175 @default.
- W4365131790 hasConceptScore W4365131790C175551986 @default.
- W4365131790 hasConceptScore W4365131790C24326235 @default.
- W4365131790 hasConceptScore W4365131790C28855332 @default.
- W4365131790 hasConceptScore W4365131790C41008148 @default.
- W4365131790 hasConceptScore W4365131790C43214815 @default.
- W4365131790 hasConceptScore W4365131790C50644808 @default.
- W4365131790 hasConceptScore W4365131790C62520636 @default.
- W4365131790 hasConceptScore W4365131790C68339613 @default.
- W4365131790 hasFunder F4320321001 @default.
- W4365131790 hasLocation W43651317901 @default.
- W4365131790 hasOpenAccess W4365131790 @default.
- W4365131790 hasPrimaryLocation W43651317901 @default.
- W4365131790 hasRelatedWork W1589376391 @default.
- W4365131790 hasRelatedWork W1967627035 @default.
- W4365131790 hasRelatedWork W2051711022 @default.
- W4365131790 hasRelatedWork W2151308525 @default.
- W4365131790 hasRelatedWork W2184497179 @default.
- W4365131790 hasRelatedWork W2378974649 @default.
- W4365131790 hasRelatedWork W2544085319 @default.
- W4365131790 hasRelatedWork W4205270981 @default.
- W4365131790 hasRelatedWork W4306177103 @default.
- W4365131790 hasRelatedWork W1506942559 @default.
- W4365131790 isParatext "false" @default.
- W4365131790 isRetracted "false" @default.
- W4365131790 workType "article" @default.